The Gifford-McMahon (G-M) type pulse tube refrigerator is a cryocooler, similar G-M refrigerators, that derives cooling from the compression and expansion of gas. However, unlike the G-M systems, in which the gas expansion work is transferred out of the expansion space by a solid expansion piston or displacer, pulse tube refrigerators have no moving parts in their cold end, but rather an oscillating gas column within the pulse tube (called a gas piston) that functions as a compressible displacer. The elimination of moving parts in the cold end of pulse tube refrigerators allows a significant reduction of vibration, as well as greater reliability and lifetime, and is thus potentially very useful in cooling cryopumps to 10 K.
G-M type pulse tube refrigerators are characterized by having a compressor that is connected to a remote expander by high and low pressure gas lines. The expander has a valve mechanism that alternately pressurizes and depressurizes the regenerators and pulse tubes to produce refrigeration at cryogenic temperatures.
G-M type pulse tube refrigerators that operate below 20 K have the disadvantage of requiring that the hot end of the pulse tube be above the cold end in order to avoid the thermal losses associated with convective circulation within the pulse tube. Conventional two-stage GM type pulse tube refrigerators typically have the valve mechanism and the hot end of the pulse tube on top. This enables the heat that is rejected at the hot end of the pulse tube to be easily transferred to the low-pressure gas and returned to the compressor where it is rejected. Conventional two stage pulse tube refrigerators also require a relatively large buffer volume(s). Two stage G-M refrigerators, which are presently being used to cool cryopumps, require no buffer volume and can be mounted in any orientation.
Most cryopumps are mounted below the vacuum chamber where space above the cryopump housing is very limited. Having the valve mechanism above the cryopump housing limits the applications of the cryopump. Thus, any options to orient the pulse tube refrigerator with the valve behind or below a cryopump housing that has a side inlet are highly desirable. Minimizing the size of the buffer volumes is also desirable.
Separating the hot end of the pulse tube from the valve introduces the problem of removing the heat that has to be rejected at the hot end of the pulse tube. From the standpoint of assembling a pulse tube refrigerator it is attractive to be able to remove the pulse tube assembly from the cryopump housing. Having the hot end of the pulse tube in vacuum and isolated from the cryopump housing makes the problem of cooling the hot end more difficult than if it is attached to the housing.
C. K. Chan, C. B. Jaco, J. Raab, E. Tward, and M. Waterman, in a paper titled “Miniature pulse tube cooler”, Proc. 7th Int'l Cryocooler Conf., Air Force Report PL-CP—93-1001 (1993) pp. 113–124, describe a Stirling single stage pulse tube that is inline, thus the hot end of the pulse tube is remote from the regenerator inlet. It has double orifice control. Heat from the hot end of the pulse tube and buffer are rejected to the base at the regenerator inlet by conduction through the buffer housing which extends the full length of the pulse tube. The hot end of the pulse tube is not attached to the vacuum housing so the entire pulse tube assembly can be easily removed.
Another method of removing heat from an inline pulse tube that is removable from a cryopump housing directs gas that is returning to the compressor to the hot end of the pulse tube where it picks up heat and transports it to the compressor to be rejected.
Gao et al., U.S. Pat. No. 5,974,807, dated Nov. 2, 1999 and entitled “Pulse Tube Refrigerator,” describes a pulse tube refrigerator capable of generating cryogenic temperatures of below 10 K that includes first and second refrigeration stages. Each stage includes a pulse tube and an associated regenerator provided at the low temperature side of the pulse tube. The high temperature ends of each pulse tube are connected by a continuous channel, while the high temperature ends of each pulse tube and the high temperature ends of each regenerator are connected by a by-pass channel. When pressure fluctuation is generated in each pulse tube at the phase difference angle of 180 degrees, respectively, a working gas is transferred between the high temperature ends of each pulse tube as controlled by an active valve, and between the high temperature ends of each pulse tube and its associated regenerator as controlled by a passive valve.
The technology disclosed by Gao et al., may be attractive for application in cryopumps because it requires no buffer volume.
Matsui et al., U.S. Pat. No. 5,845,498, dated Nov. 2, 1999 and entitled “Pulse Tube Refrigerator,” discusses the problems associated with applying prior art pulse tubes to applications where the inlet needs to be at the bottom. This patent teaches a solution to keeping the hot ends of the pulse tubes on top by using extended lengths of piping to connect components in a conventional arrangement. The volume associated with the extended tubes and the temperature patterns caused by pulse tube effects detract from the performance of this type of pulse tube.
Kawano, S. et al., U.S. Pat. No. 6,196,006, dated Mar. 6, 2001 and entitled “Pulse Tube Refrigerator” describes a pulse tube refrigerator with all of the warm gas connections being on the side of a base at ambient temperature. The regenerator is above the base; cold end on top, and the pulse tube is below the base, with the hot end fixed in the base. A long tube connects the cold end of the regenerator to the cold end of the pulse tube.
Chan, C. K. and Tward, E., U.S. Pat. No. 5,107,683, dated Apr. 28, 1992 and entitled “Multistage Pulse Tube Cooler,” describes a multistage pulse tube cooler in which a portion of the heat from each successively lower-temperature pulse tube cooler is rejected to a heat sink other than the preceding higher-temperature pulse tube cooler. This is done by having the second (and possibly third) stage pulse tube(s) reject heat at ambient temperature. This patent shows a two-stage pulse tube, each pulse tube having a single orifice and buffer volume, with the inlet to the warm regenerator at the bottom, and the hot ends of the pulse tubes on top. Furthermore the regenerators, pulse tube, and hot end heat stations are all in the vacuum space. Only the connections to the buffer tanks and the inlet tube extend through the vacuum boundary. The patent does not teach how the heat is removed from the hot end heat exchangers but the inventor's paper mentioned above describes a thermal conduction path back to the base.
Miyamoto, A. et. Al., U.S. Pat. No. 6,293,109, dated Sep. 25, 2001 and entitled “Pulse Pipe Refrigerating Machine And Cryopump Using The Same” describes a single stage pulse tube with the inlet to the regenerator at the bottom and the pulse tube oriented cold end up. The single cryopanel is a cup that is open on top. The essential teaching of this patent is the use of a working gas that at least partially condenses in the working temperature range of the cryopump. Examples are given for temperatures between 99 K and 115 K. It is known from this and other studies that the convective losses in pulse tubes when the hot end is oriented on the side or bottom become more significant as the cold station temperature is reduced below about 80 K.
The present invention optionally incorporates a number of different control concepts that have been described previously. Zhu, S. and Wu, P., “Double inlet pulse tube refrigerators: an important improvement”, Cryogenics, vol. 30 (1990), p. 514 describes the use of the second orifice and how it improves the performance of a single orifice pulse tube. A. Watanabe, G. W. Swift, and J. G. Brisson, “Superfluid orifice pulse tube below 1 K”, Advances in Cryogenic Engineering, Vol. 41B, pp. 1519–1526 (1996) describe inter-phase control. It discusses a very low temperature Stirling cycle cooler that has one passive orifice between two identical pulse tubes. J. L. Gao and Y. Matsubara, “An inter-phasing pulse tube refrigerator for high refrigeration efficiency”, in: Proceedings of the 16th International Cryogenic Engineering Conference, T. Haruyama, T. Mitsui and K. Yamafriji, ed., Eisevier Science, Oxford (1997), pp. 295–298 discuss identical dual 1, 2, and 3 stage pulse tubes with single active interconnect valves.
It is an object of this invention to provide an improved means of mounting a two-stage pulse tube refrigerator in a cryopump housing that has a side inlet.
It is another object to simplify construction of pulse tube refrigerators.
It is yet another object to improve heat rejection from the hot ends of the pulse tubes, and provide pulse tube refrigerators where the valve mechanism is below or behind the cryopump housing.